Strategic decision-making is critical to the economic success of companies and individuals. Contrasting neoclassical economics’ assumption of rational, utility-maximizing individuals, behavioral economics provides a more nuanced perspective, acknowledging the influence of cognitive and psychological biases on decision-making (Lichtenstein & Fischhoff, 1977; Tversky & Kahneman, 1974, 1981). Such deviations, so-called demand interdependencies, refer to externalities affecting an individual’s decision-making and seek to bridge the gap between economic theory and real-life strategy. Consequently, research on decision-making often encounters discrepancies between laboratory and empirical data (DellaVigna, 2009). Drawing on prospect theory, Tversky and Kahneman (1974) and DellaVigna (2009) have shown that findings from field data can differ from the standard model of a laboratory setting in three ways: nonstandard preferences, nonstandard beliefs, and nonstandard decision-making. It underscores the importance of field data to identify strategies that deviate from the standard economic model to analyze the impact of decisions. Behavioral economics analyzes frequently draw insights from professional sports, where decision-making and the respective contest outcomes are observable. Given its dynamic nature, professional basketball requires strategic decision-making from players, coaches, and teams. Success in the best basketball league worldwide, the National Basketball Association (NBA), depends on both in-game decisions, such as specific tactics, substitutions, or moves, and out-game decisions, such as trades, player signings, financial-budgetary issues, or team cohesion. In professional basketball, even minor decisions can affect team performance, as the competitive balance in the NBA is rather high. When comparing final scores, in one-third of the games, the outcome may be tied with just a two-basket difference (scoring difference ≤ 6), while in more than half of the games, achieving this requires a maximum of three baskets (points difference ≤ 9).1 This is notably surprising given the NBA’s large game count and basketball’s high-scoring nature, with teams averaging 109.2 points per game in our data. Consequently, even nuances in strategic decision-making can impact game and season outcomes. Therefore, comprehending the implications of decision making is of great relevance, not only within the sporting context of the NBA but also in broader economic and managerial settings.
This study provides insights into optimizing decision-making and enhancing team performance by examining how behavioral factors impact strategic choices in basketball. While numerous in-game decisions, such as plays and shot selection (Fichman & O’Brien, 2019; Suárez-Cadenas & Courel-Ibáñez, 2017; Suárez-Cadenas et al., 2016), substitutions (Gómez et al., 2017), tactical instructions (Christos et al., 2020; Gómez et al., 2010, 2016), or management of timeouts (Gibbs et al., 2022; Goldschmied et al., 2023) can impact success in basketball, this study focuses on a strategic decision made prior to each game: the basket choice. In the NBA, the visiting team selects which basket they prefer for the first half, determining whether they play offense in front of their bench during the first or second half.2 While the standard choice for basketball teams is to begin on offense in front of their bench, teams occasionally opt for the opposite and play on offense in front of the opponent’s bench during the first half. However, this strategic decision is the exception to the rule, possibly based on routines, assumptions, or learned information. Prospect theory suggests that such initial information can be an “anchor,” influencing subsequent judgments and estimations (Tversky & Kahneman, 1974). In basketball, this so-called anchoring bias may lead NBA teams to adhere their game management to conventional strategies, such as starting in front of their bench. Consequently, teams may make suboptimal decisions due to anchoring to biased information, perpetuating the majority behavior in basket choice.
While the few nonlaboratory field studies focus on the reasons for nonstandard decisions, little attention is paid to the strategic consequences. Hence, DellaVigna (2009) defined it as an open question of how markets react to nonstandard decisions. Furthermore, no data sets with several decisions by the same individuals are usually considered, allowing possible constant strategies to be uncovered.
However, the unobserved setting of the NBA basket choice is well suited for analyzing such strategic decisions and dynamic optimization. As the away team holds the right to choose, it offers a unique research opportunity to capture the decision and outcome. This study can enhance teams’ understanding of performance impacts and contribute to a broader comprehension of patterns in sports outcomes. Moreover, it allows for an analysis of the significance of a strategic basket choice for away teams, potentially enhancing their chances of winning and implementing management implications. This research also falls within the scope of studies on home advantage and competitive balance in sports. If the basket choice for the away team indeed boosts their likelihood of winning, an overlooked factor influencing home advantage may be rooted in the sport’s specific rules and could even diminish the established NBA home advantage.
Our findings indicate that the nonstandard choice of playing on offense in front of the opponent’s bench first increases performance and winning chances for the away team. This strategic decision is particularly advantageous when facing home teams that adhere to the standard strategy during their own away games. Similarly, those home teams that have not played together for long are also particularly susceptible to a deviation from the standard. Finally, the data indicate that the away advantage of basket choice is not impacted by the presence of spectators and can mitigate the established home advantage.
Literature Review
With basket choice in professional basketball being unexplored and without any previous findings to refer to, we rely on various dimensions that may influence on-court performance. As our empirical setting describes a decision for or against a standard choice, this section focuses on previous research on strategic decision-making and habitual behavior in sports.
The literature on habitual aspects emphasizes the significance of habits as automated and repetitive behaviors consistently performed by individuals. In the context of the sports labor market, Gorman et al. (2011) explored the recognition of patterns within performance creation among well-trained and less well-trained basketball players. Their study reveals that decision-making in sports often relies on pattern perception and the recognition of familiar play and behavioral patterns. This finding aligns with Abernethy and Russell (1987), who highlighted the faster decision-making abilities of expert athletes due to their enhanced capacity for processing information and recognizing relevant cues compared to novices. Therefore, habitual behavior acquired through perceptual patterns can stem from accumulated experience. Investigating such ingrained conduct, Lonsdale and Tam (2008) discovered a higher success rate of free throws in basketball when executed through repeated and habitual patterns, indicating that recurring habits can positively impact performance. This result is confirmed by Phelps and Kulinna (2015), who found similar outcomes in male and female high school basketball players. Additionally, Mesagno and Mullane-Grant (2010) proposed that employing preperformance routines effectively prevents performance decline in Australian football. Farrow and Abernethy (2003) extended the understanding of habitual behavior by considering the role of contextual interferences. These nonstandard disruptions, arising from new environments, situational constraints, or increased uncertainty, can hinder expert athletes’ coupling of perception and action, resulting in reduced anticipatory performance. The general effect of habitual behavior is thus revealed in such nonstandard situations. For instance, facing a left-handed opponent in tennis instead of a right-handed one deviates from the norm. This disrupts the effectiveness of learned patterns that deliberately target the opponent’s weaker hand (Loffing & Hagemann, 2012). As observed, right-handed tennis players, more accustomed to facing right-handed opponents, may have less effective strategies against left-handed opponents (Loffing et al., 2010). Similarly, left-handed volleyball players hold an advantage over right-handed opponents, who struggle to adapt to the play of left-handed players (Loffing et al., 2012). Regarding strategic decisions and playing order, Morley and Thomas (2005) revealed that a team’s captain winning the toss and electing to field first, and thus play defensively, is associated with a higher probability of winning a game in professional cricket. In a similar fashion, Kassis et al. (2021) has shown that teams whose captains can decide on the shooting sequence in soccer are more likely to win the shootout.
Research on the effects of courtside choice is limited. While Taylor et al. (2005) suggested that the direction of play can influence offensive and defensive styles in football, no conclusive findings exist. Papers on basketball identify shot selection advantages, such as shooting after offensive rebounds or the increased importance of three-point shots (Fichman & O’Brien, 2019; Suárez-Cadenas & Courel-Ibáñez, 2017; Suárez-Cadenas et al., 2016). Additionally, Gómez et al. (2010) examined general offensive and defensive strategies, demonstrating the benefits of keeping them flexible to adapt to opponents and game locations. Christos et al. (2020) emphasized the significance of offensive rebounds for an offensive strategy. Regarding in-game decisions made by coaches, Gómez et al. (2017) revealed the positive effects of newly substituted players depending on the quarter, providing strategic guidance for fielding players. Goldschmied et al. (2023) investigated the influence of timeouts on free throw performance in college basketball, suggesting that calling a timeout before a free throw can lower performance. Conversely, Gibbs et al. (2022) demonstrated that timeouts cannot halt an opponent’s run.
H1: The away team benefits from deviating from the standard, especially if the opposing team seldom does itself while playing away.
H2: Less cohesive teams are more susceptible to novel nonstandard situations.
Data and Descriptive Statistics
Data cover eight seasons of professional basketball in the NBA, spanning from 2015/2016 to 2022/2023 and comprising over 10,000 games. To ensure comparability and minimize bias from playoff games, our analysis focuses exclusively on the 9,308 regular-season games, with data sourced from https://www.basketball-reference.com.
To identify the general effect of playing with offense in front of the bench, the dummies Offense at Bench HT1 and Offense at Bench HT2 indicate whether a team is playing on offense in a given half (=1) or not (=0). Table 1 shows that most away teams (74%) opt to start games with the offense in front of their bench, a default practice in the NBA. We label this as the standard choice in the following. Consequently, the decision for Offense at Bench HT2 is the nonstandard choice, chosen in 26% of all games. Accordingly, in about one-quarter of all cases, the away team chooses to play at the basket on offense in front of the opponent’s bench first (=0) and in front of their own bench later (=1). Team-specifically, the deviation from the individual strategy, be it standard or nonstandard, is 4.1% for the entire league. For teams not exclusively committing to a single strategy within a season, this proportion rises to 11.3%. However, such deviations occur for every club, indicating that no NBA team has exclusively opted for one variant over time.
Summary Statistics
| Variable | N | M | SD | Min | Max |
|---|---|---|---|---|---|
| Points Home | 9,308 | 110.38 | 12.73 | 64 | 175 |
| Points Away | 9,308 | 108.10 | 12.85 | 68 | 176 |
| Away Win | 8,788a | 0.43 | 0.50 | 0 | 1 |
| Offense at Bench HT1 | 9,308 | 0.74 | 0.44 | 0 | 1 |
| Offense at Bench HT2 | 9,308 | 0.26 | 0.44 | 0 | 1 |
| Expected Wins Home | 9,308 | 40.79 | 10.20 | 20.50 | 67.50 |
| Expected Wins Away | 9,308 | 40.83 | 10.21 | 20.50 | 67.50 |
| Age Home | 9,308 | 26.39 | 1.82 | 21.90 | 31.40 |
| Age Away | 9,308 | 26.40 | 1.82 | 21.90 | 31.40 |
| Cohesion Home | 9,308 | 2.82 | 0.87 | 1.20 | 7.30 |
| Cohesion Away | 9,308 | 2.82 | 0.87 | 1.20 | 7.30 |
| Coach Experience Home | 9,308 | 7.72 | 6.15 | 1 | 27 |
| Coach Experience Away | 9,308 | 7.74 | 6.17 | 1 | 27 |
| Coach Tenure Home | 9,308 | 4.05 | 4.49 | 1 | 27 |
| Coach Tenure Away | 9,308 | 4.06 | 4.52 | 1 | 27 |
| Gameday Home | 9,308 | 39.54 | 22.82 | 1 | 82 |
| Gameday Away | 9,308 | 39.55 | 22.80 | 1 | 82 |
| Audience | 9,308 | 15,887.41 | 5,692.52 | 0 | 68,323 |
| Same Division | 9,308 | 0.19 | 0.39 | 0 | 1 |
| ΔPoints HT1b | 9,308 | 1.27 | 11.06 | −50 | 41 |
aGames with overtime omitted. bRegressions of Model 1 include the vector of categorial half-time scoring differences (5-point intervals).
Figure 1 shows that the frequency of the nonstandard choice has risen from approximately 22% in the 2015/2016 season to around 33% in the 2022/2023 season. This upward trajectory suggests a growing inclination among away teams to deviate from the standard choice and strategically select the opponent’s bench as their starting point. Such a development may signify a shift in teams’ strategic considerations due to dynamic optimization and an evolving approach to pregame decision-making in professional basketball.
—Development of the away teams’ decision to start with the offense in front of the opponent’s bench in %.
Citation: Journal of Sport Management 38, 5; 10.1123/jsm.2023-0305
Our primary objective in analyzing both hypotheses is to identify the direct influences of courtside selection (the away teams’ decision to select the standard or nonstandard strategy) on team performance. We initially focus on the points scored by the home and away teams in each halftime. In determining the impact of the basket choice, we investigate its consequences on the game’s outcome, specifically whether the away team wins the game. The first dependent variables we analyze for Hypotheses 1 and 2 are the Points Home and Points Away, in the respective halftime. Consistent with prior research (Pollard & Pollard, 2005; Schwartz & Barsky, 1977), the data indicate a general home advantage.
Descriptive statistics in Table 1 show a slight home advantage, with a mean home team scoring of approximately 110 points per game, while away teams average more than two points less. Since the basket choice is up to the away team, we create an associated dummy variable labeled Away Win in Table 1. However, when examining Away Win, the number of cases reduces by about 500 observations due to the exclusion of overtime games. Overtimes, played after regulations end in a draw, have specific characteristics involving fatigue, tension, stress, and other psychological demands (Gómez et al., 2015; Scanlan et al., 2019), making them a distinct scenario that can impact team performance differently and independently from the preceding game. Moreover, away teams are unlikely to consider the possibility of overtime in their strategic pregame basket choice decision.
In addition to the basket choice variable of interest, the analysis of team performance and game outcome incorporates fixed effects and several game controls on team, coach, and game level. At the team level, game controls include team quality proxied by the number of expected wins of the home and away teams. Information is derived from https://www.basketball-reference.com and details the number of expected wins as the “overrun rate for team wins during the regular season, generally set so that bets on both sides are equal” (Basketball Reference, 2023). Accordingly, Expected Wins Home and Expected Wins Away control for the expected season performance of both teams as indicated by the bookmakers. In line with numerous studies, we expect stronger teams (teams with higher numbers of expected wins) to score more points and win more likely (Mills et al., 2016; Teramoto & Cross, 2010). Further controls include team age and team cohesion. To calculate both variables, we identify the top 10 players in each squad based on minutes played in a season to avoid potential bias from those who played minimal or no part in the actual performance. Age Home and Age Away cover the average age of these 10 selected players per club and season. Younger teams generally showcase higher athleticism and growth potential and a lower likelihood of injury. In contrast, older teams leverage their players’ experience, teamwork, and gameplay understanding to excel in strategic decisions and reading opponents’ plays, resulting in better on-court execution. Cohesion Home and Cohesion Away indicate the average number of consecutive years the players have been with the club. Clubs with high cohesion may exhibit better coordination, improved communication, and a higher level of mutual trust. In high-pressure situations, cohesive teams may also be able to adapt and strategically adjust due to their familiarity and understanding of each other’s play styles (Mach et al., 2010; Montanari et al., 2008; Muthiane et al., 2015; Park et al., 2022). As expected, Table 1 demonstrates no variation within these variables between teams.
At the coach level, the analysis considers the coach’s experience and club affiliation. Coach Experience Home and Coach Experience Away cover the cumulative number of previous NBA seasons coached by a specific coach in general (Juravich et al., 2017; Roach, 2016), while Coach Tenure Home and Coach Tenure Away denote the uninterrupted number of years the coach has been with their current team. On average, coaches remain with a club for approximately 4 years. However, a notable outlier is coach Gregg Popovich of the San Antonio Spurs, who entered his 27th season with the team during the 2022/2023 season (see Table 1). A higher coaching experience or a longer tenure can lead to well-rehearsed and internalized procedures, improved tactical approaches, and influence how basket choice is handled (Frick & Simmons, 2008; James, 2007; Mielke, 2007).
The game-level variables encompass the number of gamedays for both teams, attendance, and whether both teams compete in the same division. Gameday counts the number of games played during the season, which may differ for each team due to the game schedule, and controls for potential fatigue effects (Steinfeldt et al., 2022). Audience is the number of spectators for the game, with an average attendance of 15,891 in NBA games ranging from 0 to 68,323 (c.f. Table 1). The minimum value corresponds to games during the COVID-19 pandemic when spectators were banned. We examine spectator numbers in the following, as these are more granular and can capture the COVID circumstances as well, as they correlate with these ghost games. Large audiences could potentially interfere with communication between the staff on the bench and the players on the court, influencing strategic decisions and overall team dynamics. On the other hand, supportive crowds may boost individual motivation and player performance (Goldman & Rao, 2012; Szabó, 2022). The dummy variable Same Division turns one if the home and away teams play in the same division and zero if they do not. Matchups between same-division teams are more important as both teams fight over the standings in their division (Kuehn, 2024). In estimations for second-half points, we include a set of point-difference categories after the first halftime to account for possible nonlinearities. Higher values in the first half indicate a more lopsided game in favor of the home team. It is to be expected that one-sided first halves reduce incentives and hence teams’ efforts in the second halves (Berger & Nieken, 2016; Schneemann & Deutscher, 2017). For a better interpretation of half-time differences, ΔPoints HT1 indicates that home teams lead by an average of 1.27 points at halftime.
In addition to the control variables, we test for fixed effects within the performance inferences. Season FE capture time trends in the NBA. Accordingly, rule changes or changes in the style of play may result in more or fewer points being scored (Gannaway et al., 2014; Julian & Price, 2017; Moore, 2022). Finally, we incorporate team and opponent fixed effects in our estimations. Given that Expected Wins Home and Expected Wins Away account for seasonal team strengths, Team FE and Opponent FE represent consistent team characteristics. These include factors like a club’s standing, organizational culture, and their influence on team performance and supporter support. Spectators themselves can contribute to a bigger home advantage through their level of support. Furthermore, fixed effects also encompass player development infrastructure, resources, and facilities.
Empirical Estimations
Results
Offensive Performance
Table 2 displays the estimation results from Equation (1) on the influence of the away teams’ choice on offensive performance by both teams. Home and away teams score more points when they play offense in front of their bench. Home teams score an additional 3.2 points in the first halves, while away teams’ offense improves by 2.5 points (both values compared with team performance on offense in front of the opponent’s bench). Similarly, considering the half-time score, both teams benefit from playing on offense in front of their respective benches in the second halves, with home teams improving by 2.2 points and away teams scoring an additional 3.1 points. The coefficient for home teams is greater in the first half and lower for away teams. Further testing shows no change in this effect during ghost games, indicating that communication between the coach or bench and the players on the field is not hindered by spectators or driven by no spectators, reinforcing the subconscious nature of the bench support. This suggests that the overall sense of support holds greater significance than the specific content being conveyed.
Offense at Bench Effect Per Team and Halftime
| Points | Home team | Away team | ||
|---|---|---|---|---|
| HT1 | HT2 | HT1 | HT2 | |
| Offense at Bench HT1 | 3.194*** (0.265) | 2.500*** (0.255) | ||
| Offense at Bench HT2 | 2.187*** (0.259) | 3.129*** (0.259) | ||
| Intercept | 53.881*** (2.633) | 45.863*** (2.674) | 49.337*** (2.657) | 54.068*** (2.868) |
| Controls | Yes | Yes | Yes | Yes |
| Season FE | Yes | Yes | Yes | Yes |
| Team FE | Yes | Yes | Yes | Yes |
| Opponent FE | Yes | Yes | Yes | Yes |
| N | 9,308 | 9,308 | 9,308 | 9,308 |
| R2 | .150 | .125 | .147 | .133 |
Notes. Controls for team and opponent, respectively: Expected wins, Audience, Age, Cohesion, Coach tenure, Coach experience, Gameday, Same division. Halftime 2 estimations include the vector of categorical half-time scoring differences (5-point intervals). Reference category: Team performance on offense in front of the opponent’s bench. OLS estimates, robust standard errors in parentheses. OLS = Ordinary Least Squares.
*p < .10. **p < .05. ***p < .01.
Thus, and in line with H1, it is preferable for away teams to follow the nonstandard strategy and choose to play on defense in front of their bench in the first half and to play on offense in front of their bench in the second half. Based on the results, we display the decision tree in Figure 2 with belonging path coefficients as delivered by Willoughby and Kostuk (2005) or Galariotis et al. (2018).
—Advantageousness of the basket choice for the away team.
Citation: Journal of Sport Management 38, 5; 10.1123/jsm.2023-0305
Accordingly, the away team additionally benefits from choosing to play on offense in front of the opponent’s bench in the first half, which results in playing on offense in front of its own bench in the second half. In half-time one, the away team scores 2.5 points less, while the home team suffers even more, with 3.2 points less. The point advantage for the away team is 0.694. Both teams play on offense in front of their bench in the second half and benefit, but the away team gains an additional 0.942 points over the home team. Overall, the advantage and strategy of the away team to start with their basket in front of the opponent’s bench result in a total advantage of 1.64 points. We verify this by additionally examining half-time point differentials (see Table A1 in the Appendix), which corroborate the findings. In nonstandard games, home teams score 1.19 points fewer in the first half and 1.32 points fewer in the second half. Comparing Points Home and Points Away in Table 1 reveals that home teams score, on average, 2.28 points more than the away team. Comparing games starting with the standard and the nonstandard choice, home teams score 2.72 points more in the former and only 1.05 in the latter. This suggests that the choice of basket can decisively influence the offensive performance.
Away Win
Second, we estimate the influence of basket choice on overall game outcomes. As we use the dependent variable of Away Win, we exclude games without any audience from the estimations to consider the declining home advantage as indicated by Higgs and Stavness (2021) and Leota et al. (2022). Table 3 reports in the Overall model that an away win is significantly more likely if the away team chooses the nonstandard option and starts in front of the opponent’s bench. Consequently, the basket choice translates into an additional chance to win of 3.3% for the away team in games with the nonstandard game design, as indicated by the margins. Again, this is in line with H1.
Binary Probit on Basket Choice Effect and Away Team’s Win Probability
| Away Win | Home teams | ||||||
|---|---|---|---|---|---|---|---|
| Overall | <50 | >50 | Cohesion <M | Cohesion ≥M | Coach Tenure <M | Coach Tenure ≥M | |
| Nonstandard choice | .092** (0.046) | .133** (0.054) | −.054 (0.090) | .180*** (0.062) | −.003 (0.069) | .132** (0.057) | .011 (0.079) |
| Intercept | −.701 (0.454) | −1.040* (0.555) | 1.061 (1.242) | −.334 (0.675) | −1.648** (0.741) | −.463 (0.565) | −.989 (0.955) |
| Margins | .033** | .047** | −.018 | .065*** | −.001 | .047** | .004 |
| .016 | .019 | .030 | .022 | .023 | .020 | .027 | |
| Controls | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Season FE | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Team FE | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Opponent FE | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| N | 8,232 | 6,002 | 2,230 | 4,524 | 3,708 | 5,260 | 2,972 |
| Pseudo R2 | .090 | .092 | .126 | .087 | .100 | .095 | .095 |
Notes. Controls for team and opponent, respectively: Expected wins, Audience, Age, Cohesion, Coach tenure, Coach experience, Gameday, Same division. Estimates exclude games that ended in overtime and games without audience.
aGames that ended in overtime and ghost games are excluded. OLS estimates, robust standard errors in parentheses.
*p < .10. **p < .05. ***p < .01.
As seen in Figure 1, only a minority of teams choose the nonstandard option. We suspect that teams that opt for the nonstandard choice during their away games are less affected by nonstandard choices made by opponents in home games. We further analyzed the nonstandard choice composition to detect potential selection bias relating to team strengths. First, the away team strengths do not correlate highly with the distribution of nonstandard choices (r = .062) or the proportion of nonstandard choices per team per season (r = .068). Second, a probit estimation with the dependent variable Nonstandard choice revealed a significant but small negative coefficient of −.015 (p < .001) for away team strength. This demonstrates that it is not primarily strong teams that drive the advantageousness of the nonstandard choice.
Third, Table A2 in the Appendix shows that courtside selection is more of a general strategy by teams than a game-by-game decision based on success. The baseline model (1) indicates that, while controlling for the home team’s strength, the away team’s road form, cohesion, and team fixed effects, the courtside decision from the previous game determines the same decision for the current game. Model (2) indicates that teams are more likely to stick to their previous choice if they won their previous game. This last result is in line with the idea of the hot hand (see Gilovich et al., 1985), as teams appear to stick with their strategy, especially in case of success.
Greater than 50 refers to home teams that decide for the nonstandard choice in more than 50% of their away games in a specific season. Conversely, <50 denotes teams that opt for the standard choice in more than 50% of their away games. In line with H1, the strategic nonstandard choice loses its advantage for away teams if the home team is also a >50 team. The nonstandard choice is particularly effective against <50 home teams, as the probability of an away win increases significantly by 4.7%.
Similar effects were found concerning the team characteristic of cohesion, which reflects how many seasons a club’s players have continuously played for that club. In line with H2, we expect less cohesive teams to be more susceptible to novel, nonstandard situations than teams that play together for a longer period. The dataset is divided into two groups based on the median split: home teams with cohesion above the median (M < Cohesion) and those with cohesion below the median (M > Cohesion). We suppose well-practiced home teams handle the nonstandard choice by the away team better than less well-practiced home teams. Consequently, the nonstandard choice strategy is particularly effective when away teams play against home teams that have not played together for a long time, which is in line with H3. In such cases, the probability of an away win can increase by 6.5%. While the variables <50 and M < Cohesion show somewhat comparable results, they do not correlate highly (r = .103). In contrast, coach tenure and team cohesion show a higher correlation, as new coaches often integrate new players and develop new strategies with them. Table 3 shows that teams in which a coach has not coached for a long time are also particularly susceptible to the nonstandard choice, emphasizing the importance of cohesion through overlapping coaching tenure and roster deviations.
Home and Away Advantage
As demonstrated, employing basket choice strategically can provide an advantage for the away team. Yet, it prompts consideration of how this away advantage might offset the prevalent home advantage in basketball—a phenomenon extensively substantiated by multiple authors, including a home court advantage in the NBA (Gómez & Pollard, 2011; Gómez et al., 2011; Pollard et al., 2017).
To compare the established home advantage with the away advantage showcased in the estimations above, Table 4 directly correlates the basket choice with the home or away scenario, doubling the dataset to distinguish the winning probabilities for both teams. The variables Away team and Nonstandard choice indicate the effects when the respective other equals zero. Hence, Away team shows coefficients for away teams in standard situations, while Nonstandard choice accounts for home teams in nonstandard situations. Accordingly, the coefficients of Away team confirm a significant home-court advantage in standard games (when Nonstandard choice = 0), aligning with the literature. Looking at Nonstandard choice, the coefficients indicate the negative home teams’ effects of the nonstandard choice, indicating decreasing home advantages as the Away team equaling zero here.
Binary Probit on Basket Choice Effect and General Win Probability
| Game Win | Audience | ||
|---|---|---|---|
| Overall | >0 | =0 | |
| Away team | −.427*** (0.023) | −.441*** (0.024) | −.226** (0.097) |
| Nonstandard choice | −.099*** (0.031) | −.077** (0.032) | −.418*** (0.126) |
| Away team·Nonstandard choice | .165*** (0.048) | .128*** (0.049) | .678*** (0.226) |
| Intercept | −1.086*** (0.223) | −1.028*** (0.230) | 8.679* (5.126) |
| Margins | .061*** (0.018) | .048*** (0.018) | .249*** (0.082) |
| Balanced out home advantage | 15.46% | 11.56% | 115.04% |
| Controls | Yes | Yes | Yes |
| Season FE | Yes | Yes | Yes |
| Team FE | Yes | Yes | Yes |
| N | 17,576 | 16,464 | 1,112 |
| Pseudo R2 | .059 | .062 | .075 |
Notes. Team controls: Expected wins, Audience, Age, Cohesion, Coach tenure, Coach experience, Gameday team, Same division. Estimates exclude games that ended in overtime. OLS estimates, robust standard errors in parentheses.
*p < .10. **p < .05. ***p < .01.
The Overall model in Table 4 introduces the interaction of away teams opting for the nonstandard choice in all observations to simultaneously capture the effects of the home and situational away advantages. Interaction findings corroborate Table 3, affirming the increased likelihood of winning for the away team choosing the nonstandard approach. Combining the model’s coefficients yields a remaining home advantage of .361, equivalent to a diminished away probability of −.361, still more advantageous than the standard choice.3 Hence, opting for the standard choice in general produces a coefficient of −.427; for the nonstandard choice, it is −.361. Thus, adopting the nonstandard option reduces the remaining home advantage by 15.5%. As demonstrated by Lopez et al. (2018), the NBA exhibits an overall home advantage stemming from various benefits for the home team. Figure 3 illustrates the distributions of team-specific actual and expected home win shares per season to account for potential disparities between actual and anticipated home performance. The clear correlation between anticipated and realized home wins allows for precisely capturing team-specific home advantages through Team FE and Expected Wins. In an additional regression, we observe no systematic additional home disadvantage for any team in the data and no home advantage effect on the nonstandard choice probability. Consequently, team-specific home advantages cannot account for the away team’s choice between standard and nonstandard play.
—Expected and actual share of home wins per franchise.
Citation: Journal of Sport Management 38, 5; 10.1123/jsm.2023-0305
The >0 model excludes games without an audience to prevent possible bias of the home advantage during these games (Higgs & Stavness, 2021; Leota et al., 2022). It provides a comprehensive basis for comparing home and away advantages. In games with a crowd, the nonstandard choice still offsets 11.6% of the home advantage, a notable but smaller share compared with the overall scenario. Consequently, the diminished home advantage in games without spectators predictably exerts a pronounced impact. Accounting for the individual away disadvantages (home advantages) of the Away team variable, the overall home advantage decreases by around 51% in games without spectators compared with those with attendance.4 The =0 model demonstrates that opting for the nonstandard approach in spectator-less games entirely nullifies the home advantage (115%). This implies that the reduced remaining home advantage can be offset and even overturned by the unconventional choice. Consequently, away teams were likelier to win in empty-stadium games than home teams if they began against the standard choice.
Discussion
Our study examined the consequences of an unexplored aspect: basket choice in the NBA. We observe that NBA teams demonstrate enhanced offensive performance in front of their respective benches. The coefficients on points scored and win probability showed that away teams benefit significantly from opting for the nonstandard choice, confirming Hypothesis H1 and the ideas of process disruptions made by prospect theory by Tversky and Kahneman (1974). By deliberately choosing a nonstandard strategy, the away team may exploit a strategic element of disruption and uncertainty. By unsettling the anticipated game dynamics and catching the home team off-guard, an away team’s coach can potentially induce confusion, exploit vulnerabilities, and gain an early advantage, leaving the away team in a favorable position. Hence, the results show a strategic advantage based on a single decision, similar to Maria Raya (2015), Sinnett et al. (2018), and Deutscher et al. (2023).
Our results indicate that this strategic away advantage is most potent when the opposing home team either sporadically opts for the nonstandard choice as an away team or exhibits low team cohesion in line with H2. Concerning cohesion, if the home team is not accustomed to the situation or has not played together for a long time, they may struggle to adjust to the away team’s tactics. This can lead to a slower response time in adapting to the changing dynamics of the game. Finally, we recognize that giving an away team the option to choose can mitigate the home advantage, leading to higher competitive balance, even if some home advantage effect remains, supporting the ideas of Dawson et al. (2009) or Cohen-Zada et al. (2018).
The findings of this study hold several management implications that resonate with various stakeholders. Coaches and teams should consider incorporating these findings into their game plans and tactical decisions. For instance, away teams can strategically orchestrate their offense to align with their bench in the second half, particularly when facing opponents unaccustomed to this arrangement or struggling to synchronize due to limited collective playtime. Home teams should adapt to such scenarios during their practice sessions and game strategies, enhancing their readiness for such situations. These findings underline the heightened importance of half-time adjustments, particularly as teams switch ends in the second half. Coaches need to be ready to quickly and efficiently adjust their strategies to take advantage of new opportunities or address any weaknesses revealed in the first half.
Organizers and leagues governing basketball and other sports can also derive valuable lessons. If away teams can consistently capitalize on the benefit of starting at the opponents’ bench, the home advantage effect might become less significant, which is desirable for the attractiveness and marketability of the league itself. Concerning the consumption of the NBA, more competitive games could increase fan engagement and revenue for the NBA and related businesses. The inequity stemming from home advantage or the coin toss could be directly alleviated by granting the away team certain privileges that the home team lacks. The possibility of rules balancing home-court advantage in favor of the away team may be in the NBA’s best interests as professional sports leagues seek robust competitive balance to heighten excitement within games and the league. Balanced leagues draw more spectators and enhance marketability (Humphreys, 2002; Neale, 1964). Sponsors and advertisers are economically vested, as a competitive NBA ensures substantial brand exposure and a broader audience reach. Given the notable influence on outcomes revealed by the findings, it seems fitting to engage in subsequent deliberations regarding potential alterations to rules or regulations aimed at upholding competitive equilibrium across various sports. Such discussion might further reach the sphere of the media and fans about the significance of this strategic choice and how it affects the dynamics of NBA games. It may become a talking point during pregame management of tactics and post-game discussions.
In a more general matter, our study tackles the impact of the element of random decisions on fairness. A priori coin-toss decisions, as seen in football, American football, cricket, and tennis, often impact later game outcomes considerably. Brams and Ismail (2018) emphasized that these coin-toss decisions frequently disadvantage the team losing the toss. Investigating football penalty shootouts, they find that coin tosses favor the initial kicking team and advocate for a catch-up rule to uphold equity and strategic integrity. Cohen-Zada et al. (2018) demonstrated that a specific sequence can offset the advantage of serving first in tennis. Similarly, Dawson et al. (2009) suggested that granting the weaker cricket team the choice of batting order could enhance equity and partially counteract a home-field advantage.
Subsequent studies could broaden the methodologies employed in this research to encompass alternative sports or undertake a more detailed analysis of the NBA at a finer granularity, for example, investigating whether the home team demonstrates a habituation response to the novel nonstandard situation during game-play. Furthermore, these findings might stimulate statisticians and analysts to delve deeper into the data and metrics, unearthing additional strategic benefits or trends that teams can exploit to enhance their performance.
Notes
Based on the data underlying this study.
Official playing rules 2022/2023, Rule no. 4, Section 1 a) https://ak-static.cms.nba.com/wp-content/uploads/sites/4/2022/10/Official-Playing-Rules-2022-23-NBA-Season.pdf.
Remaining home advantage = −0.427 − 0.099 + 0.165 = −0.361.
Home advantage decline during COVID = 0.226/0.441 = 0.513.
References
Abernethy, B., & Russell, D.G. (1987). Expert-novice differences in an applied selective attention task. Journal of Sport and Exercise Psychology, 9(4), 326–345.
Basketball Reference. (2023). 2022–23 NBA preseason odds. Retrieved July 2, 2023, from https://www.basketball-reference.com/leagues/NBA_2023_preseason_odds.html
Berger, J., & Nieken, P. (2016). Heterogeneous contestants and the intensity of tournaments: An empirical investigation. Journal of Sports Economics, 17(7), 631–660.
Berman, S.L., Down, J., & Hill, C.W. (2002). Tacit knowledge as a source of competitive advantage in the National Basketball Association. Academy of Management Journal, 45(1), 13–31.
Brams, S.J., & Ismail, M.S. (2018). Making the rules of sports fairer. SIAM Review, 60(1), 181–202.
Christos, K., Dimitrios, L., Christos, G., Georgios, K., & Nikolaos, S. (2020). Effect of offensive rebound on the game outcome during the 2019 basketball world cup. Journal of Physical Education & Sport, 20(6), 3651–3659.
Cohen-Zada, D., Krumer, A., & Shapir, O.M. (2018). Testing the effect of serve order in tennis tiebreak. Journal of Economic Behavior & Organization, 146, 106–115.
Dawson, P., Morley, B., Paton, D., & Thomas, D. (2009). To bat or not to bat: An examination of match outcomes in day-night limited overs cricket. Journal of the Operational Research Society, 60(12), 1786–1793.
DellaVigna, S. (2009). Psychology and economics: Evidence from the field. Journal of Economic Literature, 47(2), 315–372.
Deutscher, C., Neuberg, L., & Thiem, S. (2023). Who’s afraid of the GOATs?—Shadow effects of tennis superstars. Journal of Economic Psychology, 99, Article 102663.
Fallatah, M.I. (2021). Networks, knowledge, and knowledge workers’ mobility: Evidence from the National Basketball Association. Journal of Knowledge Management, 25(5), 1387–1405.
Farrow, D., & Abernethy, B. (2003). Do expertise and the degree of perception—Action coupling affect natural anticipatory performance? Perception, 32(9), 1127–1139.
Fichman, M., & O’Brien, J.R. (2019). Optimal shot selection strategies for the NBA. Journal of Quantitative Analysis in Sports, 15(3), 203–211.
Frick, B., & Simmons, R. (2008). The impact of managerial quality on organizational performance: Evidence from German soccer. Managerial and Decision Economics, 29(7), 593–600.
Galariotis, E., Germain, C., & Zopounidis, C. (2018). A combined methodology for the concurrent evaluation of the business, financial and sports performance of football clubs: The case of France. Annals of Operations Research, 266(1–2), 589–612.
Gannaway, G., Palsson, C., Price, J., & Sims, D. (2014). Technological change, relative worker productivity, and firm-level substitution: Evidence from the NBA. Journal of Sports Economics, 15(5), 478–496.
Gibbs, C.P., Elmore, R., & Fosdick, B.K. (2022). The causal effect of a timeout at stopping an opposing run in the NBA. The Annals of Applied Statistics, 16(3), 1359–1379.
Gilovich, T., Vallone, R., & Tversky, A. (1985). The hot hand in basketball: On the misperception of random sequences. Cognitive Psychology, 17(3), 295–314.
Goldman, M., & Rao, J.M. (2012). Effort vs. concentration: The asymmetric impact of pressure on NBA performance [Conference session]. Proceedings of the MIT Sloan Sports Analytics Conference, Boston, MA, 1–10.
Goldschmied, N., Raphaeli, M., & Morgulev, E. (2023). “Icing the shooter” in basketball: The unintended consequences of time-out management when the game is on the line. Psychology of Sport and Exercise, 68, Article 102440.
Gómez, M.A., Lorenzo, A., Ibáñez, S.J., Ortega, E., Leite, N., & Sampaio, J. (2010). An analysis of defensive strategies used by home and away basketball teams. Perceptual and Motor Skills, 110(1), 159–166.
Gómez, M.A., Lorenzo, A., Jiménez, S., Navarro, R.M., & Sampaio, J. (2015). Examining choking in basketball: Effects of game outcome and situational variables during last 5 min and overtimes. Perceptual and Motor Skills, 120(1), 111–124.
Gómez, M.A., Ortega, E., & Jones, G. (2016). Investigation of the impact of ‘fouling out’on teams’ performance in elite basketball. International Journal of Performance Analysis in Sport, 16(3), 983–994.
Gómez, M.A., & Pollard, R. (2011). Reduced home advantage for basketball teams from capital cities in Europe. European Journal of Sport Science, 11(2), 143–148.
Gómez, M.A., Pollard, R., & Luis-Pascual, J.-C. (2011). Comparison of the home advantage in nine different professional team sports in Spain. Perceptual and Motor Skills, 113(1), 150–156.
Gómez, M.A., Silva, R., Lorenzo, A., Kreivyte, R., & Sampaio, J. (2017). Exploring the effects of substituting basketball players in high-level teams. Journal of Sports Sciences, 35(3), 247–254.
Gorman, A.D., Abernethy, B., & Farrow, D. (2011). Investigating the anticipatory nature of pattern perception in sport. Memory & Cognition, 39(5), 894–901.
Graybiel, A.M. (2008). Habits, rituals, and the evaluative brain. Annual Review of Neuroscience, 31, 359–387.
Higgs, N., & Stavness, I. (2021). Bayesian analysis of home advantage in North American professional sports before and during COVID-19. Scientific Reports, 11(1), Article 14521.
Humphreys, B.R. (2002). Alternative measures of competitive balance in sports leagues. Journal of Sports Economics, 3(2), 133–148.
James, N. (2007). Coaching experience, playing experience and coaching tenure: A commentary. International Journal of Sports Science & Coaching, 2(2), 109–140.
Julian, G., & Price, J.A. (2017). Keep to the status quo: Analyzing behavioral responses to the change of ball in the NBA. International Journal of Sport Finance, 12(2), 93–108.
Juravich, M., Salaga, S., & Babiak, K. (2017). Upper echelons in professional sport: The impact of NBA general managers on team performance. Journal of Sport Management, 31(5), 466–479.
Kassis, M., Schmidt, S.L., Schreyer, D., & Sutter, M. (2021). Psychological pressure and the right to determine the moves in dynamic tournaments—Evidence from a natural field experiment. Games and Economic Behavior, 126, 278–287.
Kuehn, J. (2024). The effect of competition on the demand for skilled labor: Matching with externalities in the NBA. Journal of Economics & Management Strategy, 1–43.
Leota, J., Hoffman, D., Mascaro, L., Czeisler, M.E., Nash, K., Drummond, S.P., Anderson, C., Rajaratnam, S.M., & Facer-Childs, E.R. (2022). Home is where the hustle is: The influence of crowds on effort and home advantage in the National Basketball Association. Journal of Sports Sciences, 40(20), 2343–2352.
Lichtenstein, S., & Fischhoff, B. (1977). Do those who know more also know more about how much they know? Organizational Behavior and Human Performance, 20(2), 159–183.
Loffing, F., & Hagemann, N. (2012). Side bias in human performance: A review on the left-handers’ advantage in sports. In T. Dutta, M.K. Mandal, & S. Kumar (Eds.), Bias in human behavior (pp. 163–182). Nova Science Publishers.
Loffing, F., Hagemann, N., & Strauss, B. (2010). Automated processes in tennis: Do left-handed players benefit from the tactical preferences of their opponents? Journal of Sports Sciences, 28(4), 435–443.
Loffing, F., Schorer, J., Hagemann, N., & Baker, J. (2012). On the advantage of being left-handed in volleyball: Further evidence of the specificity of skilled visual perception. Attention, Perception, & Psychophysics, 74, 446–453.
Lonsdale, C., & Tam, J.T. (2008). On the temporal and behavioural consistency of pre-performance routines: An intra-individual analysis of elite basketball players’ free throw shooting accuracy. Journal of Sports Sciences, 26(3), 259–266.
Lopez, M.J., Matthews, G.J., & Baumer, B.S. (2018). How often does the best team win? A unified approach to understanding randomness in North American sport. The Annals of Applied Statistics, 12(4), 2483–2516.
Mach, M., Dolan, S., & Tzafrir, S. (2010). The differential effect of team members’ trust on team performance: The mediation role of team cohesion. Journal of Occupational and Organizational Psychology, 83(3), 771–794.
Maria Raya, J. (2015). The effect of strategic resting in professional cycling: Evidence from the Tour de France and the Vuelta a España. European Sport Management Quarterly, 15(3), 323–342.
Mesagno, C., & Mullane-Grant, T. (2010). A comparison of different pre-performance routines as possible choking interventions. Journal of Applied Sport Psychology, 22(3), 343–360.
Mielke, D. (2007). Coaching experience, playing experience and coaching tenure. International Journal of Sports Science & Coaching, 2(2), 105–108.
Mills, B.M., Salaga, S., & Tainsky, S. (2016). NBA primary market ticket consumers: Ex ante expectations and consumer market origination. Journal of Sport Management, 30(5), 538–552.
Montanari, F., Silvestri, G., & Gallo, E. (2008). Team performance between change and stability: The case of the Italian ‘Serie A.’ Journal of Sport Management, 22(6), 701–716.
Moore, E. (2022). Impacts of the 2018–19 NBA rule changes on scoring and the totals market. The Journal of Prediction Markets, 16(2), 39–45.
Morley, B., & Thomas, D. (2005). An investigation of home advantage and other factors affecting outcomes in English one-day cricket matches. Journal of Sports Sciences, 23(3), 261–268.
Muthiane, C.M., Rintaugu, E.G., & Mwisukha, A. (2015). The relationship between team cohesion and performance in basketball league in Kenya. International Journal of Applied Psychology, 5(4), 90–95.
Neale, W.C. (1964). The peculiar economics of professional sports. The Quarterly Journal of Economics, 78(1), 1–14.
Park, S., Kim, S., & Magnusen, M.J. (2022). Two sides of the same coin: Exploring how the bright and dark sides of team cohesion can influence sport team performance. International Journal of Sports Science & Coaching, 17(3), 519–531.
Phelps, A., & Kulinna, P. (2015). Pre-performance routines followed by free throw shooting accuracy in secondary basketball players. Biomedical Human Kinetics, 7, 171–176.
Pollard, R., & Pollard, G. (2005). Long-term trends in home advantage in professional team sports in North America and England (1876–2003). Journal of Sports Sciences, 23(4), 337–350.
Pollard, R., Prieto, J., & Gómez, M.A. (2017). Global differences in home advantage by country, sport and sex. International Journal of Performance Analysis in Sport, 17(4), 586–599.
Roach, M. (2016). Does prior NFL head coaching experience improve team performance? Journal of Sport Management, 30(3), 298–311.
Scanlan, A.T., Stanton, R., Sargent, C., O’Grady, C., Lastella, M., & Fox, J.L. (2019). Working overtime: The effects of overtime periods on game demands in basketball players. International Journal of Sports Physiology and Performance, 14(10), 1331–1337.
Schneemann, S., & Deutscher, C. (2017). Intermediate information, loss aversion, and effort: Empirical evidence. Economic Inquiry, 55(4), 1759–1770.
Schwartz, B., & Barsky, S.F. (1977). The home advantage. Social Forces, 55(3), 641–661.
Sinnett, S., Maglinti, C., & Kingstone, A. (2018). Grunting’s competitive advantage: Considerations of force and distraction. PLoS One, 13(2), Article e0192939.
Steinfeldt, H., Dallmeyer, S., & Breuer, C. (2022). The silence of the fans: The impact of restricted crowds on the margin of victory in the NBA. International Journal of Sport Finance, 2022(17), 165–177.
Suárez-Cadenas, E., & Courel-Ibáñez, J. (2017). Shooting strategies and effectiveness after offensive rebound and its impact on game result in Euroleague basketball teams. Cuadernos de Psicologıa del Deporte, 17(3), 217–222.
Suárez-Cadenas, E., Courel-Ibáñez, J., Cárdenas, D., & Perales, J.C. (2016). Towards a decision quality model for shot selection in basketball: An exploratory study. The Spanish Journal of Psychology, 19, Article E55.
Szabó, D.Z. (2022). The impact of differing audience sizes on referees and team performance from a North American perspective. Psychology of Sport and Exercise, 60, Article 102162.
Taylor, B.J., Mellalieu, D.S., & James, N. (2005). A comparison of individual and unit tactical behaviour and team strategy in professional soccer. International Journal of Performance Analysis in Sport, 5(2), 87–101.
Teramoto, M., & Cross, C.L. (2010). Relative importance of performance factors in winning NBA games in regular season versus playoffs. Journal of Quantitative Analysis in Sports, 6(3), 1–17.
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases: Biases in judgments reveal some heuristics of thinking under uncertainty. Science, 185(4157), 1124–1131.
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211(4481), 453–458.
Willoughby, K.A., & Kostuk, K.J. (2005). An analysis of a strategic decision in the sport of curling. Decision Analysis, 2(1), 58–63.
Wood, W., & Neal, D.T. (2007). A new look at habits and the habit-goal interface. Psychological Review, 114(4), 843–863.
Control Estimates on Half-Time Point Differentials
| ΔPoints | HT1 | HT2 |
|---|---|---|
| Offense at Bench | −.592** (0.291) | .662** (0.278) |
| Home | 1.681*** (0.350) | 2.820*** (0.181) |
| Offense at Bench × Home | 1.185*** (0.417) | −1.322*** (0.394) |
| Intercept | −.841 (2.446) | −4.937** (2.417) |
| Controls | Yes | Yes |
| Season FE | Yes | Yes |
| Team FE | Yes | Yes |
| Opponent FE | Yes | Yes |
| N | 18,616 | 18,616 |
| R2 | .086 | .094 |
Notes. The unit of observation is one halftime per team. Controls for team and opponent, respectively: Expected wins, Audience, Age, Cohesion, Coach tenure, Coach experience, Gameday, Same division. OLS estimates, robust standard errors in parentheses.
*p < .10. **p < .05. ***p < .01.
Away Team’s Basket Choice and Previous Road Success
| Nonstandard choice | (1) | (2) |
|---|---|---|
| Previous choice | 3.008*** (0.057) | 2.909*** (0.071) |
| Last game victory | −.100 (0.075) | |
| Previous choice × Last game victory | .258** (0.112) | |
| Intercept | −1.763*** (0.235) | −1.730*** (0.238) |
| Controls | Yes | Yes |
| Team FE | Yes | Yes |
| Opponent FE | Yes | Yes |
| N | 9,308 | 9,308 |
| R2 | .772 | .772 |
Notes. Controls: Home team’s seasonal home advantage, Away team’s current road form, Cohesion home, and away. Probit estimates, robust standard errors in parentheses.
*p < .10. **p < .05. ***p < .01.
